Melee Match-Up Chart (NTSC) [Update 008 - 09.09.28]

[Dark Hart]

btw I hate putting #'s to match-ups

 

[-ACE-]

btw I hate putting #'s to match-ups

I agree... once it gets more one-sided than say, 65/35... it's just BAD lol, I have too much trouble trying to be specific beyond that.

 

[KAOSTAR]

@blattblvd

*facepalm*

I cant believe I didnt agree b4.

Spoiler

*******!

 

[wool]

80-20 for Fox and Roy seems OK and I was in support of this a few pages back but now that I think about it more, It seems to be more 85-15 JUST because fox can camp him to death.

 

[KAOSTAR]

yea im ok with 85-15

His combo potential is limited to shorter combos and this mofo falls fast, and has ****ty recovery,

Fox is just way to fast, and 1 hit can lead to lots of damage or a stock. Dtilt fair/fsmash only goes so far.

 

[otg]

Someone want to explain why DK v. Falcon is 35/65? I could see it playing out like that, but I figure DK has a slightly better matchup then that, around 40/60 IMO. I don't feel like justifying this at all, and I could see it being one or the other, I'm just curious for peoples insight.

 

[x After Dawn x]

cause DK is ****ing huge and gets comboed like a beast because of it lol

seriously nair / uair to knee is like ****

I agree though, 35-65 is like saying DK gets ***** by Falcon as hard as Falcon gets ***** by Sheik. that ain't accurate.

 

[Gea]

The number system is stupid. Honestly the scale should go "Good luck, large disadvantage, slight disadvantage, even, slight advantage, large advantage, You best not lose."

Says much more than 40-60.

 

[DtJ Jungle]

I love losing those "best not lose" matchups

you feel great.

 

[x After Dawn x]

I love losing to Puff. Makes me feel so ****ing good that I just lost to a really, really gay character. >=(

 

[worldjem7]

guys, let's just put it at 80-20 and make him happy?

This is what this thread seems to be turning into.

So much for accuracy...

 

[Dark Hart]

This is what this thread seems to be turning into.

So much for accuracy...

lol It was just a joke, Jem. It was also a way for me to say that putting #'s to match-ups is pointless. There should only be like 5 or 7 "levels" for this chart. "Even" being the middle #, of course. It would make people happier because then we wouldn't have to worry about the difference between a 85-15 and an 80-20.

 

[worldjem7]

lol It was just a joke, Jem. It was also a way for me to say that putting #'s to match-ups is pointless. There should only be like 5 or 7 "levels" for this chart. "Even" being the middle #, of course. It would make people happier because then we wouldn't have to worry about the difference between a 85-15 and an 80-20.

Case in point.

 

[Pi]

Case in point.

Granted that this was your idea, and your thread, and you are more than welcome to run it however you see fit

But with as many variables as there are in matches; stages, characters, players, DI, etc. etc.
I'm juss not sure how feasible creating an accurate 100 point scale on which to rank characters is

And I can't really see a finalized version of it because something so minuscule might be enough to bump a matchup by 5 points in one direction or the other

I juss don't see a need for such an accurate ranking of character to character when there is very little set in stone in smash. It's not as if these numbers are going to play out time and time again with any two people.

Despite what you decide to do I would like to give you and the crew working on this chart mad props for sticking with this chart and putting so much effort into make it as accurate as it is.

I also don't want you to think that reducing the amount of points your chart runs off of would be any form of degradation. Smash is a very complex game, play styles seem to vary in almost every person you find, people have varying levels of matchup knowledge and character knowledge which in any given matchup may change the outcome. People don't need to know that a certain matchup has been estimated at a 35/65 win/lose ratio, a simple 'disadvantage' would be satisfactory.

and I'd like to say again that I appreciate the work you and the team has done on this chart

 

[otg]

I gotta be real WorldJem, making the chart simplified into 5 general categories would make this easier to read, would lead to less bickoring over trivial details like: "Is roy vs. Fox 20-80 or 15-85"? Who gives a ****? That is such a pointless detail, at the end of the day, roy will never beat a good fox, might as well make it 100-0. If you instead just take all of the horrible matchups in the game and simplify them to "Big disadvantage", then you completely eliminate all the pointless bickering.

You say you want this matchup chart to be as accurate as possible, but there is no way to do tht for the low tier matchups. Frankly, there are only a handful of tournament level low tier mains in the country (if at all), and you will NEVER create a unbiased list as a result. I think if you want a truly more generally accurate chart, we at least CONSIDER doing the chart the way the 64 boards have theirs setup.

 

[KirbyKaze]

This is what this thread seems to be turning into.

So much for accuracy...

This would bother me if we were being accurate to begin with, but we're not. The fact that we have low tier matchups included when a high level player that enters tournaments doesn't exist for over half of them (really, it's just Pikachu, Link, and maybe a few random others) makes this very inaccurate anyway.

Moreover, too many matchups are ambiguous to call it at this point with a chart like this. We've had probably six or more debates in here on whether Fox Falco is even, Fox advantage, Falco advantage and none of them ever go anywhere. Everyone has an opinion. M2K and Cactuar think Fox loses. Mango declares even. South's entirety argues Falco to be a terrible character and calls the matchup anywhere between 51-49 or 65-35 for Fox.

So as much as I'd like to believe that a chart could succeed in this format, I really don't think it can.

 

[Nintendude]

I already tried arguing this a long time ago and it was shot down, despite near universal agreement.

Maybe the creator of the SSB64 chart will start a Melee one once we finish debating the last few matchups.

 

[john!]

Changing it to 80-20 would make one person happy. Giving the chart fewer matchup levels (I think a 1-9 scale would be nice, with 5 as even) would make a lot of people happy.

 

[JPOBS]

or to make everyone happy, you could all stfu about the difference between 5 points and view this chart in your eyes.

for example, any matchup between 45-55, and 55-45 i consider even.
anything thats like 40-60 or 30-70 i consider disadvantage/advantage.

anything at 20/80 or worse is a **** matchup.

you dont need a title on the chart saying "advantage" "slight disadvantage" etc to know a reasonable range where it is.labeling the matcheups as such is just redundant.

 

[Dark Hart]

Case in point.

My point is everyone else's. Using a 0-100 scale is just not working out, and causes pointless arguments over an 85-15 and an 80-20.

 

[JPOBS]

some people say these arguments are "pointless"

and maybe thats true, but, i've found that the trend is that when soemone thinks a matchup should be canged by a few points either way, they actually get around to developing a pretty good argument for it and the matchups get discussed more intimately.

isnt that whats most important than the score given anyway?
the scores are more like bait for discussion than the result of it I've noticed.

 

[worldjem7]

or to make everyone happy, you could all stfu about the difference between 5 points and view this chart in your eyes.

for example, any matchup between 45-55, and 55-45 i consider even.
anything thats like 40-60 or 30-70 i consider disadvantage/advantage.

anything at 20/80 or worse is a **** matchup.

you dont need a title on the chart saying "advantage" "slight disadvantage" etc to know a reasonable range where it is.labeling the matcheups as such is just redundant.

This.

some people say these arguments are "pointless"

and maybe thats true, but, i've found that the trend is that when soemone thinks a matchup should be canged by a few points either way, they actually get around to developing a pretty good argument for it and the matchups get discussed more intimately.

isnt that whats most important than the score given anyway?
the scores are more like bait for discussion than the result of it I've noticed.

and this to a certain extent.

My point is everyone else's. Using a 0-100 scale is just not working out, and causes pointless arguments over an 85-15 and an 80-20.

The only time it "doesn't work out" is when several different people come in here and insist their way is right without any heed to logic or to what anyone else is saying. This is highly common when discussing low tiers. RandomScrub15341 will walk in and say "hey it's not that bad!" and expect everyone to comply, but when people throw facts and logic their way they suddenly get offended and blame the chart's format because it doesn't accommodate their unrealistic views of their character. So, in turn, they want to just generalize everything in order to make the match-up not look as bad and to make themselves feel like their character has a chance.

Melee is nowhere near a balanced fighting game.

Many characters suck, and for obvious reasons. Hiding behind some "general term" isn't going to change that, so why should the format change? "High Level Play" isn't you playing against your friends at home, it's pitting the best players from all around the continent against each other to see who will come out on top.

This thread is turning into one gigantic pity party because so many people come in here complaining about match-ups, and aren't willing to compromise. It's these people who make this format look bad when there's nothing wrong with the format in the first place.

Street Fighter, Guilty Gear, and BlazBlue, all use the same basic format, so I see no reason why it can't work for Melee.

"As accurate as possible" doesn't mean "perfect."

Obviously we're never going to get proper analysis for the low tiers, but that's to be expected because of the lack of decent players for those characters, as KirbyKaze stated. For those characters, we can only theorize as to what their match-ups are. However, this doesn't mean that any random scrub can walk in here, complain about the match-up, and expect everything to go his way, and when it doesn't meet their views, they blame the system.

If you don't know what you're talking about or have had very little tournament experience or don't place well, you can't come in here and say your low tier character does well against the high tiers because chances are even YOU don't even know what your character can and cannot do. When low tier match-ups are discussed, listen to what people are saying. Understand that the low tiers are low tiers for a reason. It's not just a fancy name given to less used characters to make you feel better when you play them. These characters are bad in many different ways and it's because they're so bad that we can't properly tell their match-ups.

Concerning high tier debates over small amounts of points: If a match-up is obviously swaying within a range of +/- 5 points, it doesn't take a genius to realize that it's best to just put that match-up in the middle. If the Falco/Fox match-up keeps swaying between 55-45 for each character, what does that tell you? That the match-up is most likely even. It's at this point that both sides should just compromise (<--Key word) and move on from that match-up. Evidently, though, no one wants to compromise, so the debate continues indefinitely.

This entire lack of compromise is very childish and needs to stop. We all play the same game, and this chart is meant to help the community that plays this game. If the community is so stubborn and only wants to bicker with itself then you end up with watered down charts like the 64 chart.

I'm trying to treat Super Smash Bros. Melee and its community as one on equal grounds as Street Fighter or Guilty Gear, but you're all showing me that it's not.

Accurate, agreeable charts are made from compromise and understanding, not from constant debate and contradiction.

 

[Pi]

Accurate, agreeable charts are made from compromise and understanding, not from constant debate and contradiction.

Heh...heh...I wish you good luck sir
when I get some more experience in matchups I'll contribute what I can
I still feel like some matchups I haven't exhausted every possible option yet

 

[joeplicate]

dude fox falco is 51.35666/48.6444
+/- 1.25%

 

[pockyD]

accuracy vs. precision

8th grade science

 

[joeplicate]

pockyd ur my smash hero

 

[idea]

compromise: make it a 1-10 scale and round everything (and debate that rounding if there's disagreement).

hmmm?

that doesn't solve the low tier matchups problem, but it makes it easier to deal with. and solves debates between whether something is even or 55-45.

also i'm starting to think the "advantage, disadvantage, etc." system wouldn't work as well for melee since there are so many more players and details and such than with 64.

 

[NES n00b]

Other fighting games can do matchups charts with matches out of 10 without much incident. Why can't we just not argue about 5 point differences? lol

 

[adumbrodeus]

Other fighting games can do matchups charts with matches out of 10 without much incident. Why can't we just not argue about 5 point differences? lol

The answer should be obvious, it's more fun that way.

 

[john!]

Heh, some people are here just to have fun and others are here to actually accomplish something.

And worldjem, I know this is your chart, but if you don't want "scrub" opinions then you should not contribute at all. Neither should I, or anyone else here (except for Kage and NES). Get the people who are actually playing at the level that we are speculating at to tell us what the matchups are, rather than having us trying to navigate abstract possibilities and sift through our own "scrub" matchup experience to contribute our personal biases, eventually turning this into a "who can post their opinion the loudest and has the best reputation" war.

Here is what needs to be done: find quality players for each character and have them give a 1-9 ranking for each matchup. Average the rankings for each matchup (from both characters' points of view) and get a number for that. E.g. Mango says Fox/Falco is 5/5, M2K says Fox/Falco is 4/6, Shiz says Fox/Falco is 6/4, so the average is 5/5, and that goes on the chart. The data pool will be much larger, but you get my point. That's really the closest we are ever going to get to approaching an accurate, unbiased chart of the highest level of play.

 

[KAOSTAR]

The number system is stupid. Honestly the scale should go "Good luck, large disadvantage, slight disadvantage, even, slight advantage, large advantage, You best not lose."

Says much more than 40-60.

I thought that originally too. But there is really no harm at all in having more detail. At this point, Id support it and say thats its better and more of a necessity for such a complex game.

Anyone with more intelligence than a pet rock should be able to look at a 95-5 matchup and say, I ****ing suck if I lose.

 

[Pi]

Idk
I guess it don't matter much
I'm still gonna play samus
I might reference this chart to make more informed counter picks though

 

[KAOSTAR]

Idk
I guess it don't matter much
I'm still gonna play samus
I might reference this chart to make more informed counter picks though

I think thats one of the best uses for this chart. Cps and just learning your character better.

 

[otg]

I dunno, I don't see how you can argue a 75/25 matchup. Like wtf does that even mean? How is that different than a 80/20, 85/15, etc etc etc? What chance is there that someone will actually even place with that character? I'm just saying, seems rather arbitrary after a certain point.

 

[SonuvaBeach]

I dunno, I don't see how you can argue a 75/25 matchup. Like wtf does that even mean? How is that different than a 80/20, 85/15, etc etc etc? What chance is there that someone will actually even place with that character? I'm just saying, seems rather arbitrary after a certain point.

This is just going in circles now.

Why are you asking redundant questions? What does 75/25 mean? That's pretty obvious. What is the difference between 80/20 and 95/5? It is the difference between Kirby vs Sheik or Fox and Yoshi or Roy. It is pretty obvious that Yoshi and Roy are better off vs top tier than Kirby and the chart reflects that.

or to make everyone happy, you could all stfu about the difference between 5 points and view this chart in your eyes.

for example, any matchup between 45-55, and 55-45 i consider even.
anything thats like 40-60 or 30-70 i consider disadvantage/advantage.

anything at 20/80 or worse is a **** matchup.

you dont need a title on the chart saying "advantage" "slight disadvantage" etc to know a reasonable range where it is.labeling the matcheups as such is just redundant.

This.

 

[otg]

Seems all made up to me, just saying. Listen I'm all for having a succesful matchup chart, just putting my 2 cents out there. You want an accurate indicator of the tiers? Weigh the matchups.

 

[N64]

Regardless of how the chart turns out to be, I support the effort. As long as it's kinda-sorta right then it at least provides some information to those interested. I still dislike the format though, and not because i'm a noob low tier that thinks my character is awesome. Pika is pretty garbage, tis unrelated.

I'm with otg for the most part. As I stated earlier my interpretation of the chart was 50-50 is even and 100-0 is marth v pichu (or insert worst matchup in the game) and then you kinda fill in everything inbetween by how **** the matchup is or isn't. I think one of the bigger issues is, yes you can mean well, but if you're trying to go for precision on an opinion-based ranking, the more precise you attempt to be, the more people will be displeased with some ranking, and will in turn voice their disagreement. There's little or no incentive for non-essential contributors to compromise at all, so in general they won't.

I can see both sides. On precision's side, IF we can get enough agreement (unlikely) and have that agreement continue as the metagame evolves, then a more precise chart can potentially provide more information. You're realistically going to have tons of arguments over it though, as some won't find it accurate enough.

On accuracy's side, you avoid a lot of debate by keeping things simple. Marth ***** Roy, but Marth REEEEALLY ***** pichu. Ok, who cares, Marth should definately win both matchups, so put it in the **** disadvantage bin and everyone's happy. If this is taken too far though, you end up with not much information at all.

As JPOBS said, you can interpret the chart how you wish by 'dumbing it down' if you dislike precision disputes, and that works fine for looking at the chart. My problem is if I'm contributing to the chart then I can't as easily 'smart it up' or make it more precise. Marth vs. pika is ****, huge disadvantage. But ask me to convert it to this chart? I dunno. Anywhere from 10-90 to 25-75. An exact number in there? Uhhhhhhhh *averages and picks one*. Now I can be quoted as putting it at a 20-80 when honestly that's just an average in where I feel relatively confident rating it. It easily could be +5 or maybe +10 in either direction.

I personally like and have supported what worldjem linked to the SF matchup chart. I have no idea if it's accurate, but I love the format. Each matchup is 3 through 7, so 3-disadvantage, 4-slightdis, 5-even, 6-slightadv, 7-advantage. Though, personally, I'd like a 1-9 chart, that seems about the right amount of precision, but 2-8 would work fine as well. Still provides plenty of information with little bickering over accuracy.

 

[KirbyKaze]

What we have right now is a very precise chart riddled with wrong information. Or information that is very controversial.

What some of us (not all of us) want, is a more general (less precise) chart that hits the ballpark better (ie. a more accurate chart). Partly because theory smash gets stupid really fast and because, frankly, some of us are tired with the "Pichu Sheik is easier than Pichu Puff", followed by "no it's not", followed by "anecdotal evidence", followed by "counter anecdotal evidence" that gets nowhere. They're both **** matchups, Pichu will never win either of them, who cares about them?

The setback to discarding the precision is mainly that it doesn't reflect the different difficulties between the top tiers and other characters with a lot of information readily available about them (who tend to have very delicate, intricate matchups with one another with a lot to consider in them) but then those are never agreed upon anyway so it's kind of moot.

 

[worldjem7]

the match-ups shown are when two characters play 10 games against each other on neutral stages and seeing how many games they'd win. The number is turned from 10 to 100 to eliminate decimals.

This is on the first post.

This is what the numbers are supposed to represent. So, 100-0 means that if two characters fought 100 games one character would win 100 and the other would win 0. If a match-up is 65-35, one character would win 65 games and the other would only win 35. This can then be equated to a percent ratio, i.e.: 65% chance of winning and 35% chance of winning.

Therefore, the difference between 65-35 and 70-30 would mean that the character with the 70 would have a slightly higher chance of winning than the character with the 65 because of X tactic or Y move or Z strategy or whatever is relevant.

The street fighter community does the same thing, as this chart was based on their system, out-of-10. Particularly, in SF4, there aren't any 8-2 or worse match-ups, so it can appear like it means "advantage/disadvantage" to someone who isn't familiar with the out-of-10 format.

I increased this to out of 100 for flexibility in precision. If a match-up isn't exactly 6-4 or isn't exactly 7-3 there was a middle ground for those kinds of disputes. It seems, though, that people are arguing more about that 0.5% than significant differences that would actually need debate in match-ups.


This is what the chart looks like if we used out-of-10:

 

[adumbrodeus]

This is on the first post.

This is what the numbers are supposed to represent. So, 100-0 means that if two characters fought 100 games one character would win 100 and the other would win 0. If a match-up is 65-35, one character would win 65 games and the other would only win 35. This can then be equated to a percent ratio, i.e.: 65% chance of winning and 35% chance of winning.

Therefore, the difference between 65-35 and 70-30 would mean that the character with the 70 would have a slightly higher chance of winning than the character with the 65 because of X tactic or Y move or Z strategy or whatever is relevant.

The street fighter community does the same thing, as this chart was based on their system, out-of-10. Particularly, in SF4, there aren't any 8-2 or worse match-ups, so it can appear like it means "advantage/disadvantage" to someone who isn't familiar with the out-of-10 format.

I increased this to out of 100 for flexibility in precision. If a match-up isn't exactly 6-4 or isn't exactly 7-3 there was a middle ground for those kinds of disputes. It seems, though, that people are arguing more about that 0.5% than significant differences that would actually need debate in match-ups.


This is what the chart looks like if we used out-of-10:

I know what the intent is, but everyone has a different interpretation of the numbers, and frankly, using it as chance ratios is ridiculous.


Primarily because it's all skewed, and most of the numbers you're dealing with are >99:<1 in terms of raw chance.


If a character has the tools to systematically beat out another character, their win ratio will always be over 80% at the top of the metagame, given an equal level of skill, and 60-40 or worse means they have the tools to systematically beat the opponent.


But the influences of skill differences are drastic in soft counters, but become less so the worse the match-up becomes. Difficulty ratio is far more useful.